Hasil Pencarian  ::  Simpan CSV :: Kembali

"Dalam perkembangan teknologi saat ini, kemampuan mesin untuk dapat belajar memiliki peranan yang sangat penting. Berbagai upaya telah dilakukan untuk mengembangkan kecerdasan buatan terhadap mesin sehingga mesin dapat melakukan pembelajaran. Salah satu macam pembelajaran mesin (machine learning) adalah dengan Brain Emotional Learning (BEL). BEL merupakan metode pembelajaran mesin yang terinspirasi dari fungsi kerja sistem limbik mamalia yang memiliki kemampuan untuk menyimpan memori, membuat keputusan dan memberi respon emosi. Dalam penerapannya, BEL telah terbukti dapat menyelesaikan berbagai masalah pembelajaran, seperti dalam masalah klasisfikasi, masalah prediksi, dan pengendalian. Pada skripsi ini, akan dilakukan perancangan dengan BEL untuk dapat mengkategorikan data melalui metode pembelajaran supervised learning dan diuji dengan data iris.Hasil pengujian menunjukkan bahwa BEL dapat digunakan untuk klasifikasi beberapa macam kelas, terdapat hubungan yang tidak linear dari faktor-faktor yang mempengaruhi proses pembelajaran terhadap hasil, konstanta β dan konstanta γ memberikan hasil akurasi rendah ketika keduanya bernilai besar, dan hasil akurasi terbaik sebesar 93,33% untuk jenis data iris. Selain itu, perbandingan dengan paper rujukan menunjukkan bahwa hasil rancangan memberikan hasil yang lebih baik daripada algoritma GDBP MLP pada epoch rendah meskipun hasil rancangan belum sebaik rujukan.

---

In todays technological development, the ability of machines to be able to learn has a very important role. Various efforts have been made to develop artificial intelligence on the machine so that the machine can do learning. One type of machine learning is with Brain Emotional Learning (BEL). BEL is a machine learning method inspired by the work function of the limbic system of mammals that has the ability to store memory, make decisions and give emotional responses. In its application, BEL has been proven to be able to solve various learning problems, such as problems in classification, prediction problems, and control. In this thesis, BEL will be designed to be able to categorize data through supervised learning methods and tested with iris data.The test results show that BEL can be used to classify several types of classes, there is a non-linear relationship of the factors that influence the learning process to results, constants and constants give low accuracy results when both are of great value, and the best accuracy results are 93, 33% for iris data types. In addition, the comparison with the reference paper shows that the design results have better results than the MLP GDBP algorithm at the lower epoch even though the design results have not been as good as the references."

"Penyakit mata berat yang telat tertangani seperti katarak, glaukoma, serta retinopati diabetik merupakan salah satu penyebab utama gangguan penglihatan dan kebutaan. Pencegahan dapat dilakukan dengan melakukan pendektesian dini melalui citra fundus. Untuk mengatasi minimnya dokter mata dan persebarannya yang masih belum merata, dilakukan pendektesian penyakit mata secara otomatis melalui gambar mata dengan pendekatan deep learning. Dalam penelitian ini, digunakan metode Transfer Learning U-Net dengan VGG16 sebagai pretrained model (V-Unet) yang telah dilatih pada online database, ImageNet. Data yang digunakan dalam penelitian ini merupakan data citra fundus yang diperoleh dari platform Kaggle. Preprocessing data pada citra fundus yang dilakukan untuk meningkatkan kinerja model adalah centered crop, resize, dan rescale. Fungsi optimasi Adam digunakan untuk meminimalkan fungsi loss ketika melatih model. Pada penelitian ini, dilakukan pemisahan data training, validasi, testing dengan 3 rasio berbeda, yaitu kasus I dengan rasio 60:20:20, kasus II dengan rasio 70:20:10, dan kasus III dengan rasio 80:10:10. Hasil penelitian ini menunjukkan bahwa V-Unet memiliki kinerja paling baik pada kasus II berdasarkan skor AUC dan running time dengan nilai rata-rata skor AUC 0,8622 dan rata-rata running time 3,7079 detik sedangkan berdasarkan nilai akurasinya V-Unet memiliki kinerja paling baik pada kasus III dengan rata-rata nilai akurasi sebesar 66,34%.

---

Untreated severe eye diseases such as cataracts, glaucoma, and diabetic retinopathy is one of the main causes of visual impairment and blindness. Prevention can be done by doing early detection through fundus images. To overcome the lack of ophthalmologists and their uneven distribution, an automatic detection of eye diseases is carried out through eye images using a deep learning approach. In this study, Transfer Learning U-Net method was used with VGG16 as a pre-trained model (V-Unet) which had been trained on the online database, ImageNet . The data used in this study is fundus image data that obtained from the Kaggle platform. Preprocessing data on the fundus image that is carried out to improve model performance is centered crop, resize, and rescale. Adam's optimization function used to minimize the loss function when training the model. In this study, the training, validation, testing data was separated with 3 different ratios, namely case I with a ratio of 60:20:20, case II with a ratio of 70:20:10, and case III with a ratio of 80:10:10. The results of this study indicate that V-Unet has the best performance in case II based on the AUC score and running time with an average AUC score of 0.8622 and an average running time of 3.7079 seconds while based on accuracy value the best case is case III with an average accuracy value of 66.34%."

"Berdasarkan Global Cancer Observatory (GLOBOCAN), tumor otak diestimasikan berada di urutan ke-19 sebagai tumor yang paling umum terjadi dan ke-12 sebagai penyebab utama kematian akibat kanker di dunia pada tahun 2020. Walaupun begitu, informasi terkait epidemiologi tumor otak di Indonesia masih sangat terbatas. Belum diwajibkannya pendataan kasus tumor di Indonesia merupakan salah satu alasannya. Tumor otak dapat dideteksi menggunakan pencitraan medis, seperti computed tomography (CT) scan dan magnetic resonance imaging (MRI). Deteksi dini tumor otak merupakan hal yang penting karena dapat meningkatkan tingkat keberlangsungan hidup dari pasien. Saat ini, banyak perkembangan teknologi yang dapat dimanfaatkan untuk membantu kehidupan manusia, salah satunya adalah deep learning (DL). Akan tetapi, data medis merupakan data yang sensitif, sehingga menjadi salah satu tantangan dalam penerapan DL di bidang kesehatan. Untuk mengatasi privasi dan keterbatasan data, terdapat metode federated learning (FL) yang memungkinkan untuk dilakukannya pelatihan data lokal pada klien tanpa menyebarkan data klien tersebut. Pada penelitian ini, akan dibentuk simulasi klasifikasi tumor otak menggunakan DL berbasis FL. Tujuan utama dari penelitian ini adalah untuk menganalisis performa model yang dihasilkan dari federated learning dan membandingkannya dengan metode training konvensional. Terdapat empat cycle data dengan tiga cycle berasal dari dataset pertama (M. Nickparvar) dan satu cycle dari dataset kedua (J. Cheng). Hasil akurasi dan F1-score tertinggi dari simulasi federated didapatkan pada epoch (jumlah putaran pelatihan data pada tiap klien) 15 dan round (jumlah putaran mulai dari tahap pembagian parameter model global kepada klien sampai dengan agregasi model) 15, yaitu 0.8375 dan 0.8384 (cycle 1), 0.7625 dan 0.7567 (cycle 2), 0.8375 dan 0.8308 (cycle 3), serta 0.7333 dan 0.7255 (dataset 2). Hasil akurasi dan F1-score tertinggi dari simulasi standard pelatihan lokal pada tiap cycle adalah 0.75 dan 0.7568 (cycle 1), 0.6875 dan 0.6677 (cycle 2), 0.675 dan 0.6744 (cycle 3), serta 0.7222 dan 0.7085 (dataset 2). Hasil akurasi dan F1-score tertinggi dari simulasi standard pelatihan all data pada tiap cycle adalah 0.7625 dan 0.7644 (cycle 1), 0.6875 dan 0.6723 (cycle 2), 0.775 dan 0.7766 (cycle 3), serta 0.6 dan 0.5355 (dataset 2). Berdasarkan pengujian hasil simulasi, korelasi epoch dan round terhadap performa model signifikan pada dataset kedua (Pacc-epoch = 0.019; PF1-epoch = 0.006; Pacc-round = 0.008; PF1-round = 0.025), tetapi hanya korelasi round yang signifikan pada dataset pertama (cycle 1 Pacc-round < 0.001 dan PF1-round < 0.001; cycle 2 Pacc-round = 0.004 dan PF1-round = 0.003; cycle 3 Pacc-round < 0.001 dan PF1-round < 0.001). Selain itu, performa model global hasil federated learning lebih baik daripada performa model lokal dan model pelatihan standard. Tidak ditemukan perbedaan signifikan antara performa model dataset pertama dengan cycle yang berbeda (Pacc between cycles = 0.679; PF1 between cycles = 0.770) serta tidak ditemukan juga perbedaan signifikan antara performa model dataset pertama dan kedua (Pacc cycle 1-dataset 2 = 0.103; Pacc cycle 2-dataset 2 = 0.334; Pacc cycle 3-dataset 2 = 0.103; PF1 cycle 1-dataset 2 = 0.140; PF1 cycle 2-dataset 2 = 0.120; Pacc cycle 3-dataset 2 = 0.140).

---

Based on the Global Cancer Observatory (GLOBOCAN), brain tumors are estimated to rank 19th among the most common tumors and 12th as the leading cause of cancer-related deaths worldwide in 2020. However, information regarding the epidemiology of brain tumors in Indonesia remains very limited. One reason is that case registration for tumors is not yet mandatory in Indonesia. Brain tumors can be detected using medical imaging, such as computed tomography (CT) scan and magnetic resonance imaging (MRI). Early detection of brain tumors is crucial as it can improve the survival rates of patients. Currently, many technological advancements can be utilized to aid human life, one of which is deep learning (DL). However, medical data is sensitive, presenting a challenge in applying DL in healthcare. To address privacy and data limitation problems, there is a method called federated learning (FL) that enables local data training on clients without sharing the clients' data. This study aims to simulate brain tumor classification using DL based on FL. The main objective of this research is to analyze the performance of the model generated from federated learning and compare it with conventional training methods. There are four data cycles, with three cycles from the first dataset (M. Nickparvar) and one cycle from the second dataset (J. Cheng). The highest accuracy and F1-score from the federated simulation were achieved at epoch (number of training rounds on each client) 15 and round (number of rounds starting from the global model parameter distribution to the clients until model aggregation) 15, which are 0.8375 and 0.8384 (cycle 1), 0.7625 and 0.7567 (cycle 2), 0.8375 and 0.8308 (cycle 3), and 0.7333 and 0.7255 (dataset 2). The highest accuracy and F1-score from the standard local training simulation in each cycle are 0.75 and 0.7568 (cycle 1), 0.6875 and 0.6677 (cycle 2), 0.675 and 0.6744 (cycle 3), and 0.7222 and 0.7085 (dataset 2). The highest accuracy and F1-score from the standard all data training simulation in each cycle are 0.7625 and 0.7644 (cycle 1), 0.6875 and 0.6723 (cycle 2), 0.775 and 0.7766 (cycle 3), and 0.6 and 0.5355 (dataset 2). Based on the simulation, the correlation between epoch and round on model performance is significant in the second dataset (Pacc-epoch = 0.019; PF1-epoch = 0.006; Pacc-round = 0.008; PF1-round = 0.025), but only the round correlation is significant in the first dataset (cycle 1 Pacc-round < 0.001 and PF1-round < 0.001; cycle 2 Pacc-round = 0.004 and PF1-round = 0.003; cycle 3 Pacc-round < 0.001 and PF1-round < 0.001). Moreover, the performance of the global model resulting from federated learning is better than the local model performance and standard all data training model performance. No significant difference was found between the performance of the first dataset with different cycles (Pacc between cycles = 0.679; PF1 between cycles = 0.770) nor between the performance of the first and second datasets (Pacc cycle 1-dataset 2 = 0.103; Pacc cycle 2-dataset 2 = 0.334; Pacc cycle 3-dataset 2 = 0.103; PF1 cycle 1-dataset 2 = 0.140; PF1 cycle 2-dataset 2 = 0.120; Pacc cycle 3-dataset 2 = 0.140)."

"Pada awal pandemi COVID-19, keputusan medis pada pasien ditentukan oleh dokter berdasarkan banyak tes medis (misalnya, tes reaksi berantai polimerase, tes suhu, CTScan atau X-ray). Metode transfer learning telah digunakan dalam beberapa penelitian dan berfokus hanya pada satu biomarker (misalnya, hanya CT-Scan atau X-Ray saja) untuk mendiagnosis pneumonia. Dalam studi terbaru, modalitas tunggal memiliki keakuratan klasifikasi sendiri dan setiap biomarker yang berbeda dapat memberikan informasi pelengkap untuk mendiagnosis COVID-19 pneumonia. Tujuan pada penelitian ini adalah membangun model multimodal yaitu dengan menggabungkan dua masukan (input) menjadi satu keluaran (output) pada tahapan pembuatan model. Dua model transfer learning yang berbeda telah digunakan pada masing-masing masukan dengan dataset open-source 2849 gambar CT-Scan dan 2849 gambar X-ray untuk mengklasifikasikan gambar CT-Scan dan gambar X-ray menjadi dua kelas: normal dan COVID-19 pneumonia. Model transfer learning yang digunakan adalah model DenseNet121, model MobileNet, model Xception, model InceptionV3, model ResNet50 dan model VGG16 untuk proses ekstraksi fitur. Alhasil, akurasi klasifikasi terbaik didapatkan sebesar 99,87% saat penggabungan jaringan ResNet50 dan VGG16. Kemudian, akurasi klasifikasi terbaik didapatkan sebesar 98,00% saat menggunakan modalitas tunggal model ResNet50 dengan data CT-Scan dan akurasi klasifikasi sebesar 98,93% untuk model VGG16 dengan data X-Ray. Metode penggabungan multimodal learning menunjukkan akurasi klasifikasi yang lebih baik dibandingkan dengan metode yang menggunakan hanya satu modalitas saja.

---

Due to COVID-19 Pandemic, medical decisions on patients were made by doctors based on many medical tests (e.g., polymerase chain reaction test, temperature test, CT-Scan or X-ray). Transfer learning methods have been used in several studies and focus on only one biomarker (eg, CT-Scan or X-Ray only) for diagnosing pneumonia. In recent studies, a single modality has its own classification accuracy and each different biomarker can provide complementary information for diagnosing COVID-19 pneumonia. The purpose of this research is to build a multimodal model by combining two inputs (inputs) into one output (output) at the modeling stage. Two different transfer learning models were used at each input with an open-source dataset of 2849 CT-Scan images and 2849 X-ray images to classify CT-Scan images and X-ray images into two classes: normal and COVID-19 pneumonia. . The transfer learning model used is the DenseNet121 model, the MobileNet model, the Xception model, the InceptionV3 model, the ResNet50 model and the VGG16 model for the feature extraction process. As a result, the best classification accuracy was obtained at 99.87% when merging the ResNet50 and VGG16 networks. Then, the best classification accuracy was obtained at 98.00% when using a single modality ResNet50 model with CT-Scan data and a classification accuracy of 98.93% for the VGG16 model with X-Ray data. The multimodal learning combination method shows better classification accuracy than the method that uses only one modality."

"Dalam beberapa tahun terakhir, keamanan data pada sistem informasi organisasi telah menjadi perhatian serius. Banyak serangan menjadi kurang terdeteksi oleh firewall dan perangkat lunak antivirus. Untuk meningkatkan keamanan, intrusion detection systems (IDS) digunakan untuk mendeteksi serangan dalam lalu lintas jaringan. Saat ini, teknologi IDS memiliki masalah kinerja mengenai akurasi deteksi, waktu deteksi, pemberitahuan alarm palsu, dan deteksi jenis serangan baru atau belum diketahui. Beberapa studi telah menerapkan pendekatan pembelajaran mesin (machine learning) sebagai solusi, dan mendapat beberapa peningkatan. Penelitian ini menggunakan pendekatan pembelajaran ensemble (ensemble learning) yang dapat mengintegrasikan manfaat dari setiap algoritma pengklasifikasi tunggal. Pada penelitian ini, dibandingkan tujuh pengklasifikasi tunggal untuk mengidentifikasi pengklasifikasi dasar yang digunakan untuk model ensemble learning. Kemudian dataset IDS terbaru dari Canadian Institute for Cybersecurity yaitu CSE-CIC-IDS2018 digunakan untuk mengevaluasi model ensemble learning. Hasil percobaan menujukan bahwa implementasi metode ensemble learning khususnya majority voting dengan tiga algoritma dasar (gradient boosting, decision tree dan logistic regression) dapat meningkatkan nilai akurasi lebih baik dibandingkan implementasi algoritma klasifikasi tunggal, yaitu 0,988. Selanjutnya, implementasi teknik pemilihan fitur spearman-rank order correlation pada dataset CSE-CIC-IDS2018 menghasilkan 23 dari 80 fitur, dan dapat meningkatkan waktu pelatihan model, yaitu menjadi 11 menit 4 detik dibanding sebelumnya 34 menit 2 detik.

---

In recent years, data security in organizational information systems has become a serious concern. Many attacks are becoming less detectable by firewall and antivirus software. To improve security, intrusion detection systems (IDSs) are used to detect anomalies in network traffic. Currently, IDS technology has performance issues regarding detection accuracy, detection times, false alarm notifications, and unknown attack detection. Several studies have applied machine learning approaches as solutions. This study used an ensemble learning approach that integrates the benefits of each single classifier algorithms. We made comparisons with seven single classifiers to identify the most appropriate basic classifiers for ensemble learning. Then the latest IDS dataset from the Canadian Institute for Cybersecurity, CSE-CIC-IDS2018, was used to evaluate the ensemble learning model. The experimental results show that the implementation of the ensemble learning method, especially majority voting with three basic algorithms (gradient boosting, decision tree and logistic regression) can increase the accuracy rate better than the implementation of a single classification algorithm, which is 0.988. Furthermore, the implementation of the spearman-rank order correlation feature selection technique in the CSE-CIC-IDS2018 dataset produced 23 of the 80 features, and could increase the model training time, which was 11 minutes 4 seconds compared to 34 minutes 2 seconds before."

"ABSTRAKBerdasarkan Kementerian Pendidikan dan Kebudayaan Indonesia pada tahun 2019, memperkirakan bahwa 70% dari anak berkebutuhan khusus belum dilayani. Luasnya Negara Republik Indonesia dan pandemi covid-19 saat ini serta adanya kebijakan pemerintah belajar dari rumah agar tetap terpenuhi hak anak serta mempertahankan kualitas hidup AbK-CP, maka kebutuhan pendamping terlatih sangat mendesak. Pelatihan dengan pembelajaran daring menjadi solusi terbaik dalam perluasan dan pemerataan pelayanan di Indonesia. Cerebral Palsy (CP) pada anak adalah gangguan dalam perkembangan yang dapat mempengaruhi kemampuan motorik dan menyeimbangkan tubuh yang memiliki tingkat keparahan. Material dalam penelitian ini pendamping AbK khususnya CP, gawai, aplikasi teleconference dengan materi pembelajaran berupa video maupun slide. Data diambil dengan penilaian observasi kinerja pendamping, akurasi dilakukan oleh tenaga ahli. Studi ini menggunakan analisis dengan metode kualitatif deskriptif untuk mengevaluasi kinerja hasil pelatihan bagi pendamping AbK-CP secara daring. Berdasarkan metode uji korelasi Spearmen nilai <0,05 didapatkan pada evaluasi capaian, proses pengulangan serta profesi responden yang menunjukkan bahwa ada korelasi yang signifikan. Secara kuantitatif keberhasilan pencapaian kinerja, baik berkaitan dengan pengulangan maupun berkaitan dengan profesi, mencapai 90%. Pelatihan dengan pembelajaran daring ini dapat memberikan layanan pada AbK-CP sesuai dengan karakteristik kebutuhan yang spesifik, oleh pendamping terlatih dengan metode ramah Ramah Otak Ramah HatI (RORI), serta dapat dilakukan dirumah.

---

ABSTRACTChildren with Special Needs Education-Cerebral palsy (CSN-CP) based on the Indonesian Ministry of Education and Culture in 2019, predicted that 70% of SNC had not been served. The extent of the Republic of Indonesia geographic area and pandemic Covid-19 nowadays and also the Government's policy of learning from home in order to remain to fulfill the children's rights and maintain the quality of CSN-CP life, so the needs of trained caregiver is very urgent. Training with online learning is the best solution in the expansion and equality service in Indonesia. Cerebral palsy (CP) in children is a disorder in development that can affect the motor ability and balance the body that has its severity. The material in this study caregiver CSN especially CP, gadgets, application of teleconference with the digital learning materials in the form of video and slides. Data is taken with an observation assessment of the caregiver performance; experts do accuracy. This study uses analysis with a qualitative descriptive method to evaluate the performance of training results for caregiver CSN-CP online. Based on the correlation test method Spearmen the value of < 0.05 is obtained in the evaluation of the achievement, the repetition process as well as the profession of respondents indicating that there is a significant correlation. Quantitatively, the success of performance, both related to repetition and related to the profession, reaches 90%. Training with online learning can provide services to CSN-CP in accordance with specific needs characteristic, by trained caregiver with friendly method of friendly brain-friendly heart (RORI), and can be done at home."

"Thoroughly examining brain-behavior relationships in atypically developing children, this important volume integrates theories and data from multiple disciplines. Leading authorities present research on specific clinical problems, including autism, Williams syndrome, learning and language disabilities, ADHD, and issues facing infants of diabetic mothers. In addition, the effects of social stress and maltreatment on brain development and behavior are reviewed. Demonstrating the uses of cutting-edge methods from developmental neuroscience, developmental psychology, and cognitive science, the contributors emphasize the implications of their findings for real-world educational and clinical practices. Illustrations include eight pages in full color."

"Brain-based learning allows trainers and educators to promote long-term retention in learners and create a fun, interactive atmosphere. Brain-based learning uses our knowledge of how the brain processes and stores information to create more effective learning programs. This Infoline introduces readers to the concept of brain-based learning and provides tips for how to use the concepts in training design and delivery."